Since the beginning of July my Opteron box was randomly crashing and
being rebooted by hardware watchdog.  Today it finally did it in front
of me, and this patch will hopefully fix it.

The problem is that at the end of June (the 28th, to be exact: commit
47f176fd, "[PATCH] Using msleep()
instead of HZ") rtc_get_rtc_time was converted to use msleep() instead
of busy waiting.  But rtc_get_rtc_time is used by hpet_rtc_interrupt,
and scheduling is not allowed during interrupt.  So I'm reverting this
part of original change, replacing msleep() back with busy loop.

The original code was busy waiting for up to 20ms, but on my hardware in
the worst case update-in-progress bit was asserted for at most 363
passes through loop (on 2GHz dual Opteron), much less than even one
jiffie, not even talking about 20ms.  So I changed code to just wait
only as long as necessary.  Otherwise when RTC was set to generate
8192Hz timer, it stopped doing anything for 20ms (160 pulses were
skipped!) from time to time, and this is rather suboptimal as far as I
can tell.
Signed-off-by: Petr Vandrovec <vandrove@vc.cvut.cz>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

We have found what seems to be a small bug in __vm_enough_memory() when
sysctl_overcommit_memory is set to OVERCOMMIT_NEVER.

When this bug occurs the systems fails to boot, with /sbin/init whining
about fork() returning ENOMEM.

We hunted down the problem to this:

The deferred update mecanism used in vm_acct_memory(), on a SMP system,
allows the vm_committed_space counter to have a negative value.

This should not be a problem since this counter is known to be inaccurate.

But in __vm_enough_memory() this counter is compared to the `allowed'
variable, which is an unsigned long.  This comparison is broken since it
will consider the negative values of vm_committed_space to be huge positive
values, resulting in a memory allocation failure.

Signed-off-by: <Jean-Marc.Saffroy@ext.bull.net>
Signed-off-by: <Simon.Derr@bull.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

tcp_write_xmit caches the cwnd value indirectly in cwnd_quota.  When
tcp_transmit_skb reduces the cwnd because of tcp_enter_cwr, the cached
value becomes invalid.

This patch ensures that the cwnd value is always reread after each
tcp_transmit_skb call.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

MSS changes can be lost since we preemptively initialize the tso_segs count
for an SKB before we %100 commit to sending it out.

So, by the time we send it out, the tso_size information can be stale due
to PMTU events.  This mucks up all of the logic in our send engine, and can
even result in the BUG() triggering in tcp_tso_should_defer().

Another problem we have is that we're storing the tp->mss_cache, not the
SACK block normalized MSS, as the tso_size.  That's wrong too.
Signed-off-by: David S. Miller <davem@davemloft.net>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This avoids the whole #ifdef mess by just getting a copy of
dentry->d_inode before d_delete is called - that makes the codepaths the
same for the INOTIFY/DNOTIFY cases as for the regular no-notify case.
I've been running this under a Gnome session for the last 10 minutes.
Inotify is being used extensively.
Signed-off-by: John McCutchan <ttb@tentacle.dhs.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

When recently addressing remarks by Alexey Dobriyan about
the isp116x-hcd, I introduced a bug in the driver. Please
apply the attached patch to fix it.
Signed-off-by: Olav Kongas <ok@artecdesign.ee>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This patch has a one line oops fix, plus related cleanups.

 - The bugfix uses microframe scheduling data given to the hardware to
   test "is this a periodic QH", rather than testing for nonzero period.
   (Prevents an oops by providing the correct answer.)

 - The cleanup going along with the patch should make it clearer what's
   going on whenever those bitfields are accessed.

The bug came about when, around January, two new kinds of EHCI interrupt
scheduling operation were added, involving both the high speed (24 KBytes
per millisec) and low/full speed (1-64 bytes per millisec) microframe
scheduling.  A driver for the Edirol UA-1000 Audio Capture Unit ran into
the oops; it used one of the newly supported high speed modes.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The patch which went in was correct, but not quite what I had in mind.
Here is a patch to update that a little bit. Original patch is at:
 http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=4749f32da939d4e4160541b2cadc22492bb507ecSigned-off-by: Pete Zaitcev <zaitcev@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

In yenta_socket, we default to using the resource setting of the CardBus
bridge.  However, this is a PCI-bus-centric view of resources and thus needs
to be converted to generic resources first.  Therefore, add a call to
pcibios_bus_to_resource() call in between.  This function is a mere wrapper on
x86 and friends, however on some others it already exists, is added in this
patch (alpha, arm, ppc, ppc64) or still needs to be provided (parisc -- where
is its pcibios_resource_to_bus() ?).
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Some PCI devices (e.g. 3c905B, 3c556B) lose all configuration
(including BARs) when transitioning from D3hot->D0.  This leaves such
a device in an inaccessible state.  The patch below causes the BARs
to be restored when enabling such a device, so that its driver will
be able to access it.

The patch also adds pci_restore_bars as a new global symbol, and adds a
correpsonding EXPORT_SYMBOL_GPL for that.

Some firmware (e.g. Thinkpad T21) leaves devices in D3hot after a
(re)boot.  Most drivers call pci_enable_device very early, so devices
left in D3hot that lose configuration during the D3hot->D0 transition
will be inaccessible to their drivers.

Drivers could be modified to account for this, but it would
be difficult to know which drivers need modification.  This is
especially true since often many devices are covered by the same
driver.  It likely would be necessary to replicate code across dozens
of drivers.

The patch below should trigger only when transitioning from D3hot->D0
(or at boot), and only for devices that have the "no soft reset" bit
cleared in the PM control register.  I believe it is safe to include
this patch as part of the PCI infrastructure.

The cleanest implementation of pci_restore_bars was to call
pci_update_resource.  Unfortunately, that does not currently exist
for the sparc64 architecture.  The patch below includes a null
implemenation of pci_update_resource for sparc64.

Some have expressed interest in making general use of the the
pci_restore_bars function, so that has been exported to GPL licensed
modules.
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Revert this June 17 patch: it broke persistence of timers across execve().

Cc: Roland McGrath <roland@redhat.com>
Cc: george anzinger <george@mvista.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The IA32 ptrace emulation currently returns the wrong registers for fs/gs;
it's returning what x86_64 calls gs_base.  We need regs.gsindex in order
for GDB to correctly locate the TLS area.  Without this patch, the 32-bit
GDB testsuite bombs on a 64-bit kernel.  With it, results look about like
I'd expect, although there are still a handful of kernel-related failures
(vsyscall related?).
Signed-off-by: Daniel Jacobowitz <dan@codesourcery.com>
Acked-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

We had a user whose apps weren't working correctly because his "rtc" wasn't
working fully.

For the sake of simplicity, it seems sensible to always enable HPET RTC
emulation.

Remove a special config option for HPET_EMULATE_RTC and make it directly
depend on HPET_TIMER and RTC. This will avoid the hangs when EMULATE_RTC
is not configured and when some userlevel script depends on RTC interrupt,
as in:

http://bugzilla.kernel.org/show_bug.cgi?id=4904Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

mremap's move_vma is applying __vm_stat_account to the old vma which may
have already been freed: move it to just before the do_munmap.

mremapping to and fro with CONFIG_DEBUG_SLAB=y showed /proc/<pid>/status
VmSize and VmData wrapping just like in kernel bugzilla #4842, and fixed by
this patch - worth including in 2.6.13, though not yet confirmed that it
fixes that specific report from Frank van Maarseveen.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The included patch fixes a problem where a inotify client would receive a
delete event before the file was actually deleted.  The bug affects both
dnotify & inotify.
Signed-off-by: John McCutchan <ttb@tentacle.dhs.org>
Signed-off-by: Robert Love <rml@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The inotify help text still refers to the character device.  Update it.

Fixes kernel bug #4993.
Signed-off-by: Robert Love <rml@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The attached patch makes sure that a keyring that failed to instantiate
properly is destroyed without oopsing [CAN-2005-2099].

The problem occurs in three stages:

 (1) The key allocator initialises the type-specific data to all zeroes. In
     the case of a keyring, this will become a link in the keyring name list
     when the keyring is instantiated.

 (2) If a user (any user) attempts to add a keyring with anything other than
     an empty payload, the keyring instantiation function will fail with an
     error and won't add the keyring to the name list.

 (3) The keyring's destructor then sees that the keyring has a description
     (name) and tries to remove the keyring from the name list, which oopses
     because the link pointers are both zero.

This bug permits any user to take down a box trivially.
Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The attached patch prevents an error during the key session joining operation
from hanging future joins in the D state [CAN-2005-2098].

The problem is that the error handling path for the KEYCTL_JOIN_SESSION_KEYRING
operation has one error path that doesn't release the session management
semaphore. Further attempts to get the semaphore will then sleep for ever in
the D state.

This can happen in four situations, all involving an attempt to allocate a new
session keyring:

 (1) ENOMEM.

 (2) The users key quota being reached.

 (3) A keyring name that is an empty string.

 (4) A keyring name that is too long.

Any user may attempt this operation, and so any user can cause the problem to
occur.
Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The kexec boot is not successful on some power machines since all CPUs are
getting removed from global interrupt queue (GIQ) before kexec boot.  Some
systems always expect at least one CPU in GIQ.  Hence, this patch will make
sure that only secondary CPUs are removed from GIQ.
Signed-off-by: Haren Myneni <hbabu@us.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This code was never designed to handle more than one instance of do_work()
running at once.
Signed-Off-By: Alasdair G Kergon <agk@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Acked-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Signed-off-by: Jim Keniston <jkenisto@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The recent change to never ignore the bitmap, revealed that the bitmap isn't
begin flushed properly when an array is stopped.

We call bitmap_daemon_work three times as there is a three-stage pipeline for
flushing updates to the bitmap file.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Firstly, R1BIO_Degraded was being set in a number of places in the resync
code, but is never used there, so get rid of those settings.

Then: When doing a resync, we want to clear the bit in the bitmap iff the
array will be non-degraded when the sync has completed.  However the current
code would clear the bitmap if the array was non-degraded when the resync
*started*, which obviously isn't right (it is for 'resync' but not for
'recovery' - i.e.  rebuilding a failed drive).

This patch calculated 'still_degraded' and uses the to tell bitmap_start_sync
whether this sync should clear the corresponding bit.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The code currently will ignore the bitmap if the array seem to be in-sync.
This is wrong if the array is degraded, and probably wrong anyway.  If the
bitmap says some chunks are not in in-sync, and the superblock says everything
IS in sync, then something is clearly wrong, and it is safer to trust the
bitmap.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Until the bitmap code was added,

   modprobe md

would load the md module.  But now the md module is called 'md-mod', so we
really need an alias for backwards comparability.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

no_overlay bttv parameter implemented to fix OOPS on some PCI chipsets
(like some VIA) with these behaviors:

1) If pci_quicks does identify the chip as having troubles to
   handle PCI2PCI transfers, no_overlay defaults to 1. The user may force
   it to 0, to reenable (not recommended).

2) For newer chipsets not blacklisted, no_overlay=1 is provided as a
   workaround until PCI chipset included on /drivers/pci/quirks.c

Thanks to Bodo Eggert <7eggert@gmx.de>
Signed-off-by: Michael Krufky <mkrufky@m1k.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

CONFIG_KEXEC breaks UP builds because of a misspelled smp_release_cpus().
Also, the function isn't defined unless built with CONFIG_SMP but it is
needed if we are to go from a UP to SMP kernel.  Enable it and document it.

Thanks to Steven Winiecki for reporting this and to Milton for remembering
how it's supposed to work and why.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Patch fixes oops caused by ide interfaces not on pci.  pcibus_to_node
causes the kernel to crash otherwise.  Patch also adds a BUG_ON to check if
hwif is NULL.
Signed-off-by: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Patch from Michael Gernoth

As discussed on the handhelds.org Jornada mailinglist, I take over
maintainership of the currently unmaintained Jornada 720-port in
the mainline kernel.
Signed-off-by: Michael Gernoth <michael@gernoth.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

Several people noticed we dropped quite a bit on benchmark figures.
OK, it was my fault but unfortunately I discovered I ran out of brown
paper bags a while ago and forgot to reorder them.

The issue is that a construct introduced in the conversion of the
driver to use the transport class keyed off whether the block request
was tagged or not.  However, the aic7xxx driver doesn't properly set
up the block layer TCQ (it uses the wrong API), so the driver now
things all requests are untagged and we keep it to a queue depth of a
single element.  Oops.

The fix is to use the correct TCQ API.
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>

Fix up arm26, cris, frv, m68k, parisc and sh64 too..

x86_64 had hardcoded the VM_ numbers so it broke down when the numbers
were changed.
Signed-off-by: Alexander Nyberg <alexn@telia.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This fixes five bugs in the key management syscall interface:

 (1) add_key() returns 0 rather than EINVAL if the key type is "".

     Checking the key type isn't "" should be left to lookup_user_key().

 (2) request_key() returns ENOKEY rather than EPERM if the key type begins
     with a ".".

     lookup_user_key() can't do this because internal key types begin with a
     ".".

 (3) Key revocation always returns 0, even if it fails.

 (4) Key read can return EAGAIN rather than EACCES under some circumstances.

     A key is permitted to by read by a process if it doesn't grant read
     access, but it does grant search access and it is in the process's
     keyrings. That search returns EAGAIN if it fails, and this needs
     translating to EACCES.

 (5) request_key() never adds the new key to the destination keyring if one is
     supplied.

     The wrong macro was being used to test for an error condition: PTR_ERR()
     will always return true, whether or not there's an error; this should've
     been IS_ERR().
Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-Off-By: Linus Torvalds <torvalds@osdl.org>

This removes the calls to device_suspend() from the shutdown path that
were added sometime during 2.6.13-rc*.  They aren't working properly on
a number of configs (I got reports from both ppc powerbook users and x86
users) causing the system to not shutdown anymore.

I think it isn't the right approach at the moment anyway.  We have
already a shutdown() callback for the drivers that actually care about
shutdown and the suspend() code isn't yet in a good enough shape to be
so much generalized.  Also, the semantics of suspend and shutdown are
slightly different on a number of setups and the way this was patched in
provides little way for drivers to cleanly differenciate.  It should
have been at least a different message.

For 2.6.13, I think we should revert to 2.6.12 behaviour and have a
working suspend back.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>